A Field Study of Related Video Recommendations: Newest, Most Similar, or Most Relevant?

Yifan Zhong, Tahir Lazaro Sousa Menezes, Vikas Kumar, Qian Zhao, F Maxwell Harper

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Many video sites recommend videos related to the one a user is watching. These recommendations have been shown to influence what users end up exploring and are an important part of a recommender system. Plenty of methods have been proposed to recommend related videos, but there has been relatively little work that compares competing strategies. We describe a field study of related video recommendations, where we deploy algorithms to recommend related movie trailers. Our results show that recency- and similarity-based algorithms yield the highest click-through rates, and that the recency-based algorithm leads to the most trailer-level engagement. Our findings suggest the potential to design non-personalized yet effective related item recommendation strategies.
Original languageEnglish (US)
Title of host publicationProceedings of the 12th ACM Conference on Recommender Systems
PublisherACM
Pages274-278
Number of pages5
ISBN (Print)978-1-4503-5901-6
DOIs
StatePublished - 2018

Publication series

NameProceedings of the 12th ACM Conference on Recommender Systems

Fingerprint

Light trailers
Recommender systems

Keywords

  • field study
  • item similarity
  • movie trailers
  • recommender systems
  • related item recommendations

Cite this

Zhong, Y., Menezes, T. L. S., Kumar, V., Zhao, Q., & Harper, F. M. (2018). A Field Study of Related Video Recommendations: Newest, Most Similar, or Most Relevant? In Proceedings of the 12th ACM Conference on Recommender Systems (pp. 274-278). (Proceedings of the 12th ACM Conference on Recommender Systems). ACM. https://doi.org/10.1145/3240323.3240395

A Field Study of Related Video Recommendations: Newest, Most Similar, or Most Relevant? / Zhong, Yifan; Menezes, Tahir Lazaro Sousa; Kumar, Vikas; Zhao, Qian; Harper, F Maxwell.

Proceedings of the 12th ACM Conference on Recommender Systems. ACM, 2018. p. 274-278 (Proceedings of the 12th ACM Conference on Recommender Systems).

Research output: Chapter in Book/Report/Conference proceedingChapter

Zhong, Y, Menezes, TLS, Kumar, V, Zhao, Q & Harper, FM 2018, A Field Study of Related Video Recommendations: Newest, Most Similar, or Most Relevant? in Proceedings of the 12th ACM Conference on Recommender Systems. Proceedings of the 12th ACM Conference on Recommender Systems, ACM, pp. 274-278. https://doi.org/10.1145/3240323.3240395
Zhong Y, Menezes TLS, Kumar V, Zhao Q, Harper FM. A Field Study of Related Video Recommendations: Newest, Most Similar, or Most Relevant? In Proceedings of the 12th ACM Conference on Recommender Systems. ACM. 2018. p. 274-278. (Proceedings of the 12th ACM Conference on Recommender Systems). https://doi.org/10.1145/3240323.3240395
Zhong, Yifan ; Menezes, Tahir Lazaro Sousa ; Kumar, Vikas ; Zhao, Qian ; Harper, F Maxwell. / A Field Study of Related Video Recommendations: Newest, Most Similar, or Most Relevant?. Proceedings of the 12th ACM Conference on Recommender Systems. ACM, 2018. pp. 274-278 (Proceedings of the 12th ACM Conference on Recommender Systems).
@inbook{cdaf4185dea54357a374e7a39e5d3264,
title = "A Field Study of Related Video Recommendations: Newest, Most Similar, or Most Relevant?",
abstract = "Many video sites recommend videos related to the one a user is watching. These recommendations have been shown to influence what users end up exploring and are an important part of a recommender system. Plenty of methods have been proposed to recommend related videos, but there has been relatively little work that compares competing strategies. We describe a field study of related video recommendations, where we deploy algorithms to recommend related movie trailers. Our results show that recency- and similarity-based algorithms yield the highest click-through rates, and that the recency-based algorithm leads to the most trailer-level engagement. Our findings suggest the potential to design non-personalized yet effective related item recommendation strategies.",
keywords = "field study, item similarity, movie trailers, recommender systems, related item recommendations",
author = "Yifan Zhong and Menezes, {Tahir Lazaro Sousa} and Vikas Kumar and Qian Zhao and Harper, {F Maxwell}",
year = "2018",
doi = "10.1145/3240323.3240395",
language = "English (US)",
isbn = "978-1-4503-5901-6",
series = "Proceedings of the 12th ACM Conference on Recommender Systems",
publisher = "ACM",
pages = "274--278",
booktitle = "Proceedings of the 12th ACM Conference on Recommender Systems",

}

TY - CHAP

T1 - A Field Study of Related Video Recommendations: Newest, Most Similar, or Most Relevant?

AU - Zhong, Yifan

AU - Menezes, Tahir Lazaro Sousa

AU - Kumar, Vikas

AU - Zhao, Qian

AU - Harper, F Maxwell

PY - 2018

Y1 - 2018

N2 - Many video sites recommend videos related to the one a user is watching. These recommendations have been shown to influence what users end up exploring and are an important part of a recommender system. Plenty of methods have been proposed to recommend related videos, but there has been relatively little work that compares competing strategies. We describe a field study of related video recommendations, where we deploy algorithms to recommend related movie trailers. Our results show that recency- and similarity-based algorithms yield the highest click-through rates, and that the recency-based algorithm leads to the most trailer-level engagement. Our findings suggest the potential to design non-personalized yet effective related item recommendation strategies.

AB - Many video sites recommend videos related to the one a user is watching. These recommendations have been shown to influence what users end up exploring and are an important part of a recommender system. Plenty of methods have been proposed to recommend related videos, but there has been relatively little work that compares competing strategies. We describe a field study of related video recommendations, where we deploy algorithms to recommend related movie trailers. Our results show that recency- and similarity-based algorithms yield the highest click-through rates, and that the recency-based algorithm leads to the most trailer-level engagement. Our findings suggest the potential to design non-personalized yet effective related item recommendation strategies.

KW - field study

KW - item similarity

KW - movie trailers

KW - recommender systems

KW - related item recommendations

UR - http://doi.acm.org/10.1145/3240323.3240395

UR - http://www.mendeley.com/research/field-study-related-video-recommendations-newest-most-similar-most-relevant

U2 - 10.1145/3240323.3240395

DO - 10.1145/3240323.3240395

M3 - Chapter

SN - 978-1-4503-5901-6

T3 - Proceedings of the 12th ACM Conference on Recommender Systems

SP - 274

EP - 278

BT - Proceedings of the 12th ACM Conference on Recommender Systems

PB - ACM

ER -